Vol. 30: 209-215,1997 DISEASES OF AQUATIC ORGANISMS Dis Aquat Org

Published September 30

Differential phagocytic ability of the circulating haemocyte types of the carpet shell clam Ruditapes decussatus (Mollusca: Bivalvia)

Carmen ~ o p e z l, Maria J. Carballal l, Carlos ~ z e v e d o ~ , Antonio villalbal

'Centre of Marine Research, Ministry of Fisheries and Aquaculture of Galicia, POB 208, E-36600 Vilagarcia de Arousa, Pontevedra. Spain

'Department of Cell Biology, Institute of Biomedical Sciences. University of Porto, P-4000 Porto. Portugal

ABSTRACT: Phagocytosis of foreign materials by haemocytes is an important aspect of the internal defence of bivalve molluscs. Two main haemocyte types can be distinguished in the haemolymph of the clam Ruditapes decussatus: granulocytes and hyalinocytes. The ability of clam haemocytes to phago- cytose zymosan particles, Vibrio P1 cells and trophozoites of the protistan parasite Perkinsus atlant~cus was demonstrated by means of in vitro assays. However, clam haemocytes did not phagocytose atlanticus zoospores in the assays. Granulocytes showed the highest phagocytic capacity in each assay. Phagocytic capacity of haemocytes was not significantly affected by clam age. An ultrastructural study of phagocytosis showed the following sequence of events: engulfment of particles by pseudopods, for- mation of a phagocytic vacuole, fusion of lysosomes with the phagocytic vacuole, and digestion of the particles giving rise to residual bodies that might be discharged.

KEY WORDS: Ruditapes decussatus . Haemocytes . Phagocytosis . Vibrio P1 . Perkinsus atlanticus

INTRODUCTION

It is well-known that haemocytes are involved in the immune system of bivalve molluscs. Bivalve haemo- cytes are usually classified into 2 types, granulocytes and hyalinocytes (Cheng 1975, Auffret 1988). Never- theless, heterogeneous cell populations are not pre- cisely characterized in terms of their respective de- fence functions.

Phagocytosis of non-self materials by haemocytes is one of the defence mechanisms of bivalve molluscs (Feng 1988). It is known that haemocytes are the phagocytic cells in bivalve molluscs.

Infectious diseases are a very important problem in mollusc aquaculture. Among bivalve molluscs com- mercial losses occur as a result of mortalities caused by the presence of protozoa, rickettsia and chlamydia, and bacteria and viruses (Lauckner 1983). The avail- ability of pathogen purification protocols (Mialhe et al. 1985, 1988, Le Gall & Mialhe 1992) has permitted

the study of different aspects of defence mechanisms against a number of pathogens.

Several authors have studied the phagocytosis pro- cess and associated cytotoxic mechanisms in species belonging to the Veneridae family: e.g. Mercenaria mercenaria (Anderson 1994) and Ruditapes decussatus (Lopez et al. 1994). R. decussatus is one of the most important species of this family in Europe, from a com- mercial point of view. Populations of this species are affected by 2 important epizootics caused by Perkinsus atlanticus (Azevedo 1989) and Vibrio P1 (Paillard et al. 1994).

Morphological characterization of the circulating haemocytes of Ruditapes decussatus was studied as a first step of research on cellular defence and 2 cell types were found, granulocytes and hyalinocytes (Lopez et al. 1997).

The purpose of the present study was to evaluate the phagocytic capacity of the haemocyte types of Rudi- tapes decussatus through in vitro assays using live

0 Inter-Research 1997 Resale of full article not permitted

210 Dis Aquat Org 30: 209-215, 1997

cells of the clam pathogens mentioned above and inert particles as phagocytosis stimulants.

MATERIALS AND METHODS

Haernolymph collection. Haemolyrnph was taken from the posterior adductor muscle of clams Ruditapes decussatus using a syringe, through a hole made in the shell margin. It was collected diluted 1:3 in a mod- ified anti-aggregant Alsever solution (MAS) (20.80 g 1-' glucose, 8.00 g 1-' sodium citrate, 3.36 g 1-' EDTA, 22.50 g 1-' sodium chloride in distilled water). Haemo- lymph of 5 specimens was pooled and the number of haemocytes was estimated with a Malassez haemo- cytometer.

For phagocytosis assays, the concentration in MAS was reduced to 3 % by centrifugation, supernatant removal, and resuspension of haemocytes in sterile filtered seawater (SFSW).

Phagocytosis assays. Light microscopy. In phagocy- tosis assays zymosan, Vibrio P1 bacteria (VPl), tropho- zoites and zoospores of Perkinsus atlanticus were used as phagocytosis stimulants.

In the in vitro phagocytosis assays, 2 X 105 haemocytes were incubated in suspension with different particle stimulants for 1 h at room temperature (Fries & Tripp 1980), in Eppendorf tubes. The particle-to-haemocyte ratio was 5:l for zymosan, trophozoites and zoospores of Perkinsus atlanticusand 20:l for VP1. After the incu- bation, cytospin slides were performed, fixed, stained with rapid Hemacolor kit (Merck) and mounted.

The percentage of haemocytes corresponding to each cell type in the cytospins was estimated after examination of 300 cells by light microscopy.

A similar in vitro assay, with zymosan as phagocyto- sis stimulant, was performed from 2 groups of 15 clams of different ages (31.00 + 0.31 mm and 23.93 * 0.27 mm in longest axis) to test the influence of age on phago- cytic capacity.

Phagocytosis assays. Electron microscopy. Similar in vitro assays were performed to study ultrastructural aspects of the process of phagocytosis. The number of haemocytes in these assays was 3 X 106 and zymosan and VP1 were used as phagocytosis stimulant. No stage of Perkinsus atlanticus was used in these assays. After incubation, the suspension was fixed with 2.5 % glutaraldehyde in PIPES buffer (0.1 M, at pH 7.2 and 7 % sucrose) for 1 h at 4"C, and centrifuged (750 X g, 10 min). The pellet was washed in PIPES buffer with sucrose for 2 h at 4'C, and post-f~xed in 1 % osmium tetroxide in the buffer for 75 min at 4OC. The cells were then embedded in 1.5 % agar at 40°C, and quickly cen- trifuged (1700 X g, 5 min). The pellet was dehydrated and embedded in Epon resin. Ultrathin sections (50 to

70 nm) were stained with uranyl acetate and lead citrate and examined in a transmission electron micro- scope (Jeo1100CXII).

Production of Perkinsus atlanticus zoospores. Gills of clams which were heavily infected with P. atlanticus were incubated in fluid thioglycollate medium (FTM) at room temperature, in the dark, for 3 d (Ray 1966). Gills were then removed from FTM, blotted on filter paper, transferred to a Petri dish with SFSW containing antibiotics (Penicillin G 500 units ml-' and strepto- mycin 500 mg ml-l) and chopped with a scalpel. After attachment of the zoosporangia to the base of Petri dish, the supernatant was poured into another dish (carefully, to prevent the detachment of zoosporangia). Then, the first Petri dish was filled with SFSW with antibiotics and this was repeated in the other Petri dishes. Petri dishes were kept at 28°C and it was pos- sible to see zoospores in the SFSW on the third day. Zoospores were observed to be actively motile just before the performance of in vitro phagocytosis assays.

Purification of trophozoites of Perkinsus atlanticus. The following procedure was adapted from the proce- dure for the purification of Bonamia ostreae from Ostrea edulis (Mialhe et al. 1988).

Only gills from Ruditapes decussatus heavily in- fected with Perkinsus atlanticus were used in the purification. Clams that were heavily infected could be distinguished by naked eye because white pustules were visible in the gills and mantle. Up to 12 clams were used in each purification.

Gills were homogenised with FSW with the addition of 0.5 % Tween 80 (FSWT). The homogenate was then sieved progressively through 100, 75 and 20 pm nylon meshes. The resultant filtrate was spun at 2500 X g for 30 min at 8'C. The pellet was suspended in FSWT and layered onto a cushion of 15% sucrose in FSWT. After centrifugation, the supernatant was discarded and the pellet diluted in FSWT and layered onto a gradient of sucrose in FSWT (35% and 15%), and centrifuged again. Trophozoites were found at the Interface 15%/35%. This fraction was resuspended in FSWT and spun, and the pellet was resuspended in 1 m1 of FSWT and layered onto a Percoll gradient (10, 20, 30, 40 and 50%), containing 0.5 M NaCl to prevent osmotic shock in trophozoites. This gradient was spun and trophozoites were found at the interfaces between 20%/30% and 30%/40%. These fractions were col- lected with a syringe. An equivalent volume of FSW was added and the solution layered onto a cushion of 15% sucrose in FSW (to remove the Percoll) and centrifuged. The supernatant was discarded and the trophozoites which were found in the pellets were resuspended in FSW and centrifuged as a final rinse. Vibrio P1 culture. The strain Vibrio P1 used in this

study were provided by the IFREMER laboratory (La

Lopez et al.: Phagocytic capacity of clam haernocyles 211

Tremblade, France). The bacteria were maintained in marine agar (Difco) at 4°C. For phagocytosis assays VP1 was seeded in marine broth (Difco). The number of VP1 was determined by measuring the optical den- sity (1.25 OD is equivalent to 1500 X 10bVP1 ml-'). The marine broth was eliminated by centrifugation (2000 X

g, 15 min); VPl was resuspended in FSW and mea- sured again.

Zymosan. Preparation of a suspension of yeast cell walls using zymosan A (Sigma, USA) followed that described by Dikkeboom et al. (1987).

RESULTS

Phagocytosis of zymosan assays

Both granulocytes and hyalinocytes showed phago- cytic capacity in the in vitro phagocytosis assays with zymosan, although this capacity was higher in granulocytes. Of total haemocytes, 49% phagocytosed zymosan particles (Fig. 1). Of this population, 72.4% of the acidophilic granulocytes and 40.9% of the baso- philic granulocytes had internalized zymosan particles at the end of the assays, whereas only 3.4% of the hyalinocytes had internalized particles. In most of the cells that had phagocytosed, 1 or 2 zymosan particles were observed regardless of cell type. Higher numbers of phagocytosed particles (up to 8) were occasionally observed.

The ultrastructural study allowed observation of entire zymosan particles inside primary phagosomes and partially digested particles in secondary phago- somes (Fig. 2). The number of cytoplasmic granules was lower in granulocytes with internalized zymosan, compared to those granulocytes without phagocytosed zymosan. This difference was probably due to the fusion of granules with primary phagosomes after zymosan internalization. In advanced stages of the phagocytosis process, haemocytes showed a vacuo- lated appearance with glycogen particles in the cyto- plasm and highly degraded material in secondary phagosomes. Images of residual material discharge from haemocytes could be observed.

Phagocytosis of VP1 assays

In this case, 55.3% of the haemocytes had phago- cytosed bacteria Vibrio P1 at the end of the assay. As in the previous case, each haemocyte type was able to phagocytose bacteria. Acidophilic granulo- cytes appeared as the most phagocytic (76.6% of them internalized bacteria) followed by basophilic granulocytes (39.7 %) and hyalinocytes (33.0%). Up to

10 phagocytosed bacteria were observed in a single cell (Fig. 3).

Electron microscopy confirmed the capacity of haemocytes to phagocytose VP1 bacteria. Partially de- graded VP1 bacteria were observed inside secondary phagosomes (Fig. 4), and rarely inside primary phago- somes. Haemocytes with phagocytosed bacteria showed some vacuoles with residual materials. The presence of small vesicles detached from the cytoplasmic mem- brane was very common in phagocytic haemocytes.

Phagocytosis of trophozoites and zoospores of Perkinsus atlanticus

It was not possible to distinguish the haemocyte type after trophozoites had been internalized because of the large size of the latter. However, it was possible to dis- tinguish the haemocyte type in the trophozoite engulf- ment stage. Internalized trophozoites were found in 43.7% of the haemocytes at the end of the assays. At the engulfment stage, 41.1 % of the acidophilic granu- locytes, 37.2% of the basophilic granulocytes and 19.0% of the hyalinocytes were found to be engaged in the process.

In contrast, clam haemocytes were not able to phago- cytose zoospores of Perkinsus atlanticus in the in vitro assays.

Influence of clam age in phagocytic capacity

Table 1 shows the percentages of haemocytes with phagocytosed Zymosan in 2 groups of different age clams. The difference found between the groups was not significant.

DISCUSSION

The results of in vitro assays showed that all clam haemocyte types had phagocytic capacity, although granulocytes showed a greater ability. These findings are consistent with results reported from Crassostrea

Table 1. Percentage of haemocyte types of Ruditapes decus- satus with phagocytosed zymosan particles corresponding to 2 groups of clams of different ages. Mean size * standard error corresponding to each clam group are indicated in

parentheses

Haernocyte type Older clams Younger clams (31.00 * 0.31 mm) (23.93 + 0.27 mm)

Hyalinocytes 4.71 % 8.81 % Basophhc granulocytes 38.89 % 41.18% AcidophiLic granulocytes 74 68 % 73.08 % Total 47.78% 51.33%

Lopez et al.: Phagocytic capacity of clam haemocytes 213

Figs. 1 to 4. Ruditapes decussatus haemocytes. Brightfield micrograph haemocytes after exposure to zymosan (Zy) par- ticles. (~1400) . - Ultrastructure of a granulocyte with phagocytosed zymosan particles. N: nucleus; Z,: zymosan; Pp: primary phagosome; Sp: secondary phagosome; GR: granule. (~5800) . Fig. 3. Bnghtfield micrograph of haemocytes after exposure to Vibrio P1 (VP1) (~1400) . Fig. 4. Ultrastructure of a granulocyte with phagocytosed Vibrio P1 (VP1) bacterium within secondary

p h a g G e (S,). N: nucleus; GR: granule; G: golgl complex; mi: mitochondria. (x15 200)

virginica and Mercenana mercenaria by Foley & Cheng (1975) and Renwrantz et al. (1979) from C. virginica. In contrast, Ruddell (1971) indicated that in C. virginica the agranular amebocytes appeared to have greater phagocytic ability. However, Hawkins & Howse (1982) for C, virginica and Auffret (1986) for Ostrea edulis found that only the granulocyte type was phagocytic. Among granulocytes, the acidophilic granulocytes of Ruditapes decussatus were found to be more phago- cytic, which is in agreement with the observations of Pipe et al. (1994). The higher phagocytic capacity observed in R. decussatus granulocytes is consistent with the occurrence of higher levels of hydrolytic enzymes in these cells (Lbpez et al. in press). The func- tion of bivalve hyalinocytes is not known but it is pos- sible that they have another specific function different to phagocytosis. Hose et al. (1990) and Clare & Lumb (1994) indicated that crustacean hyalinocytes have a role in the clotting process. Bivalves do not have humoral clotting factors but Suzuki et al. (1991) indi- cated that agranular haemocytes of pearl oyster Pinc- tada fucata show the ability to form a cellular sheath and produce an extracellular matrix in the wound healing process.

Around 50 % of Ruditapes decussatus haemocytes showed phagocytic ability in our assays. Fries & Tripp (1980) noticed phagocytic ability in 100% of Merce- naria mercenaria haemocytes. Differences could be due to experimental procedures or to intrinsic proper- ties of the bivalve species. La Peyre et al. (1995) reported a significant difference in the percentage of phagocytic haemocytes between Crassostrea gigas and C, virginica using the same experimental proce- dure with both species.

Ruditapes decussatus haemocytes showed the ability to phagocytose all stimulant particles used in the assays, except zoospores of Perkinsus atlanticus. In contrast, haemocytes of Crassostrea virginica are able to phagocytose zoospores of P. marinus. Different ex- perimental conditions should be used to further study the inability of clam haemocytes to phagocytose P. atlanticus zoospores. Feng (1966, 1967) and Feng & Stauber (1968) found that Haplosporidiun nelsoni, Hexamita nelsoni and Staphylococus aureus induced little or no phagocytosis by C, virginica haemocytes. Some differences were found in the percentage of phagocytic hyalinocytes depending on the phago-

cytosis stimulant used. However, the percentage of phagocytic granulocytes was similar in all the assays. La Peyre et al. (1995) found a significant difference in the percentage of phagocytic haemocytes between assays using P. marinus trophozoites and those using zymosan.

According to Cheng (1983) and Renwrantz (1990), the phagocytic process can be subdivided into 4 steps: (1) chemotaxis, involving the attraction of the phago- cyte to the non-self material, (2) recognition, involving the attachment of the foreign material to the surface of the phagocyte, (3) internalization, and (4) intracellular degradation. Light and electron rnicroscopy studies of the phagocytic process by clam haemocytes demon- strated internalization and degradation but chemotaxis and recognition phases were not considered.

Internalization of foreign particles by clam haemo- cytes occurred by invagination of the cell surface followed by pseudopod formation and particle inter- nalization into endocytotic vacuoles (primary phago- some). A similar model was described by Cheng (1975). Two more models of the same event were described by Bang (1961) and Renwrantz et al. (1979).

Our fine structural observations indicate that intra- cellular degradation of internalized material occurs in secondary phagosomes of the clam haemocytes. In the phagocytosis assays, fusion between granules and the primary phagosome was not observed, but haemocytes with secondary phagosomes contained fewer granules. It is possible that secondary phagosomes result from the fusion of lysosomes with primary phagosomes. Auffret (1986) suggested that the formation of sec- ondary phagosomes results from the fusion of granules and primary phagosomes in Ostrea edulis. Neverthe- less the loss of granules from granulocytes of Ruditapes decussatus during the in vitro phagocytosis assays could also be due to the release of lysosomal enzymes from granulocytes into the serum (degranulation). A degranulation process associated with phagocytosis was reported by Foley et al. (1977) in Mercenana mercenaria and Hinsch & Hunte (1990) in Crassostrea virginica.

In this phase, lysosomal enzymes play a very impor- tant role. Mohandas et al. (1985) and Oubella et al. (1994) demonstrated that challenge with bacteria pro- duced a hypersynthesis of lysosomal enzymes by haemocytes. A number of lysosomal enzymes were de-

214 Dis Aquat Org 30: 209-215,1997

tected in Ruditapes decussatus haemocytes and the presence of acid phosphatase in the granules suggests that these granules are lysosomes (Lopez et al. in press). The presence of lysosomal enzymes in bivalve mollusc haemocytes was cited by several authors (Moore & Gelder 1985, Feng 1988, Beckrnann et al. 1992). The absence of oxidative metabolism coupled to phagocytosis in R. decussatus haemocytes was previ- ously reported (L6pez et al. 1994).

Cheng (1981) described 2 models of intracellular degradation of phagocytosed materials in bivalve haemocytes. One of them involved diffusion of mole- cules through the phagosomal membrane into the cytoplasm after enzyme digestion. Glucose is then polymerized into glycogen in the cytoplasm. Indi- gestible material remains as residual bodies which are eventually discharged from the cell. The other model suggests digestion of bacteria beginning within a primary phagosome. The material is transferred from primary to secondary phagosomes and glycogen granules are synthesized from glucose inside these phagosomes. Then, the secondary phagosomal wall disintegrates and the glycogen granules are freed into the cytoplasm.

In our study, transference of digested material from primary to secondary phagosomes was not detected. However, glycogen granules were detected inside sec- ondary phagosomes. Discharged residual bodies were also observed in Ruditapes decussatus haemocytes.

Acknowledgements. We thank the team of Eric Mialhe (IFREMER, La Tremblade, France) for supplying the Vibrio P1 bacteria. The clams for this study were supplied by Manuel Franco. This study was partially supported by a scholarship of Caixa de Galicia in 1993. for C.L. during a stay at the IFREMER laboratory.

LITERATURE CITED

Anderson RS (1994) Hemocyte-derived reactive oxygen inter- mediate production in four bivalve mollusks. Dev Comp Immunol 18(2):89-96

Auffret M (1986) Internal defence in b~valve molluscs: ultra- structural observations on the fate of experimentally in- jected bacterial in Ostrea edulis granular hemocytes. In:. Vivares CP, Bonami JR, Jaspers E (eds) Pathology in marine aquaculture. European Aquaculture Society, Bre- dene, Special Pub1 9, p 351-356

Auffret M (1988) Bivalve hemocyte morphology. In: Fisher WS (ed) Disease processes in marine bivalve molluscs. Ameri- can Fisheries Society Special Publication 18, Bethesda, MD. p 169-177

Azevedo C (1989) Fine structure of Perkinsus atlanticus n. sp. (Apicomplexa, Perkinsea) parasite of the clam Rudtapes decussatus from Portugal. J Parasitol 75(4):627-635

Bang FB (1961) Reaction to injury in the oyster (Crassostrea virginica). Biol BuU (Woods Hole) 121(1):57-68

Beckman N, Morse MP. Moore Chl(1992) Comparative study

of phagocytosls In normal and diseased hemocytes of the bivalve Mya arenaria. J Invertebr Pathol59:124-132

Cheng TC (1975) Functional morphology and biochemistry of molluscan phagocytes. Am NJ Acad Sci 266:343-379

Cheng TC (1981) Bivalves. In: Ratcliffe NA, Rowley AF (eds) Invertebrate blood cells. Academic Press, London, p 233-300

Cheng TC (1983) Internal defense mechanisms of molluscs against invadlng microorganisms: personal remin~scences. Trans Am Microsc Soc 102(3):185-193

Clare AS, Lumb G (1994) Identification of haemocytes and their role in clotting in the blue crab, CalLinectes sapidus. Mar Biol 118:601-610

Dikkeboom R, Tijnagel JMGH, Mulder E, Knaap van der WPW (1987) Hemocytes of the pond snail Lymnaea stag- nalis generate reactive forms of oxygen. J Invertebr Pathol 49:321-331

Feng S Y (1966) Experimental bacterial infection in the oyster Crassosfrea virginica. J lnvertebr Pathol 8:505-511

Feng SY (1967) Responses of molluscs to foreign bodies, with special reference to the oyster. Fed Proc 26(6):1685-1692

Feng SY (1988) Cellular defence mechanisms of oysters and mussels. In: Flsher WS (ed) Disease processes in marine bivalve molluscs. American Fisheries Society Special Pub- lication 18, Bethesda, MD, p 153-168

Feng SY, Stauber LA (1968) Experimental hexamitiasis in the oyster Crassostrea virginica. J Invertebr Pathol 10:94-110

Foley DA, Cheng TC (1975) A quantitative study of phago- cytosis by hemolymph cells of the pelecypods Crassostrea virginica and Mercenaria mercenaria. J Invertebr Pathol 25:189-197

Foley DA, Cheng TC (1977) Degranulation and other changes of molluscan granulocytes associated with phagocytosis. J Invertebr Pathol 29:321-325

Fries CR, Tripp MR (1980) Depression of phagocytosis in Mercenaria mercenaria following chemical stress. Dev Comp Immunol4:233-244

Hawkins EW. Howse RD (1982) Ultrastructure of cardiac hemocytes and related cells in the oyster Crassostrea vir- ginica. Trans Am Microsc Soc 101(3):241-252

Hinsch GW, Hunte M (1990) Ultrastructure of phagocytosis by hemocytes of the American oyster. In: Perkins FO, Cheng TC (eds) Pathology in marine science. Academic Press, San Diego, p 479-488

Hose JE, Martin GG, Gerard AS (1990) A decapod hemocyte classification scheme integrating morphology, cytochem- istry, and function. Biol Bull (Woods Hole) 178:33-45

La Peyre JF. Chu FLE, Meyers JM (1995) Haemocytic and humoral activities of eastern and Pacific oysters following chaUenge by the protozoan Perkinsus marinus. Fish & Shellfish lmmunol 5:179-190

Lauckner G (1983) Diseases of Mollusca: Bivalvia. In: Kinne 0 (ed) Diseases of marine animals, Vol 11. Biologische Anstalt Helgoland, Hamburg, p 477-961

Le Gall G, Mialhe E (1992) Purification of rickettsiales-like organisms associated with Pecten maximus (Mollusca: Bivalvia) serological and blochemica1 characterization. Dis Aquat Org 12:215-220

Lopez C, Carballal MJ, Azevedo C, Villalba A (1997) Morphological characterization of the hernocytes of the clam, Rud~tapes decussatus (Mollusca: Bivalvia). J Inver- tebr Path01 69:51-57

Lopez C, Carballal MJ, Azevedo C, Villalba A (in press) Enzyme characterization of the circulating haemocytes of the carpet shell clam, Ruditapes decussatus (Mollusca: Bivalvia), in relation to defence function. Fish & Shellfish Immunol

Lopez et al.. Phagocytic capacity of clam haemocytes 215

Lopez C, V~llalba A, Bachere E (1994) Absence of generation of active oxygen radicals by the hemocytes of the clam, Ruditapes decussatus (Mollusca: Bivalvia) coupled with phagocytosis. J Invertebr Pathol 64:188-192

Mialhe E, Bachere E, Chagot D, Grizel H (1988) Isolation and purification of the protozoan Bonamia ostreae (Pichot et al. 1980). a parasite affecting the flat oyster Ostrea edulis L. Aquaculture 71:293-299

Mialhe E , Bachere E , Le Bec C, Grizel H (1985) Isolement et purification de h/larteilia (Protozoa: Ascetospora), para- sites de bivalves marins. C R Acad Sci Paris Ser I11 301. 137-142

Mohandas A. Cheng TC, Cheng J B (1985) Mechanism of lysosomal enzyme release from Mercenaria rnercenaria granulocytes: a scanning electron microscope study. J Invertebr Pathol 45:189-197

Moore CA, Gelder SR (1985) Demonstration of lysosomal enzymes in hemocytes of Mercenaria mercenaria (Mollusca: Bivalvia). Trans Am Microsc Soc 104(3): 242-249

Oubella R, Paillard C, Maes P, Auffret M (1994) Changes in hemolymph parameters in the Manila clam Ruditapes philippinarum (Mollusca: Bivalvia) following bacterial challenge. J lnvertebr Pathol 64:33-38

Edjtonal responsibiljty: Albert Sparks, Seattle, Washington, USA

Palllard C, Maes P, Oubella R (1994) Brown nng disease in clams. Annu Rev Fish Dis 4:219-240

Pipe R K , Holden J A , Farley SR (1994) The separation and characterisation of haernocytes from the marine mussel, Mytllus edulis. Dev Comp Immunol (S1):113

Ray SM (1966) A review of a culture method for detecting Dermocystidium marinurn with suggested modifications and precautions. Proc Natl Shellfish Assoc 54:55-80

Renwrantz L (1990) Internal defense system of Mytilus edulis. In: Stefano GB (ed) Neurobiology of Mytilus edulis. Studies in Neuroscience, No. 2, Manchester University Press. Manchester, p 257-275

Renwrantz L, Yoshino T, Cheng T, Auld K (1979) Size deter- mination of hemocytes from the American oyster Cras- sostrea virginica, and the description of a phagocytosis mechanism. Zoo1 Jb Physiol83:l-12

Ruddell CL (1971) Elucidation of the nature and function of the granular oyster amebocytes through histochemical studies of normal and traumatized oyster tissues. Histo- cheme 26:98-112

Suzuki T. Yoshinaka R, Mizuta S, Funakoshi S. Wada K (1991) Extracellular matrix formation by amebocytes during epithelia1 regeneration in the pearl oyster Pinctada fucata. Cell Tissue Res 266:75-82

Manuscript received: September 9, 1996 Revlsed versjon accepted: May 20, 1997